Jacob Weisberg has written an annoying piece about fighting povery in the aftermath of Katrina that is straight out of Slate’s “everything you know is wrong!” handbook. Rather than fighting attempts to turn the disaster zone into a conservative utopia of low taxes and weak regulations, Weisberg says that liberals should go along with Bush:

[A]ll of us have a stake in an experiment that tells us whether conservative anti-poverty ideas, uh, work. If the conservative war on poverty succeeds, even in partial fashion, we will all be better for its success. And if it fails, we will have learned something important about how not to fight poverty.

Unfortunately, Ezra Klein’s Tapped post responding to Weisberg is an ugly mix of invective and know-nothingism:

[Weisberg’s] argument, basically, is that liberals should let conservatives do whatever they wish to New Orleans because America needs to figure out whether Republican solutions to poverty will succeed — and if they don’t, it’s because they’re scared the conservative plans will work. The people of New Orleans, we can only assume, will not be told they’re participating in an economic version of the Tuskegee experiments; we’ll just let them get buffeted about out of sheer respect for the scientific method. Testing hypotheses is much more important than a few thousand black folk.

Weisberg, bizarrely, seems to think a liberal unwillingness to make New Orleans into a Grover Norquist fantasyland stems from a fear that it’ll work. Au contraire, it stems from a fear that it won’t work, and that scores of poor people who just had their lives demolished will now be further punished by a regionally restricted Gilded Age. Liberals do not think this because they are afraid, they think this because they are liberals. If they did not believe conservative solutions would fail, they would be conservatives. And if either group, liberals or conservatives, are so unsure of their policies that they believe antithetical programs should be applied for the experimental value of it, they should really exit the debate with all possible speed — these are real people we’re talking about, they shouldn’t be subdivided into control groups and experiment zones.

Klein is of course correct that liberals should not support policies that they believe will fail solely as an experiment. But the rest of his argument is horribly misguided.

First, the comparison to the notorious Tuskegee experiments, in which poor African American men with syphillis were left untreated for years, is absurd. Presumably, conservatives are proposing measures that they believe will help people in the Gulf, not withholding proven medical treatments from people with a disease.

In addition, Klein’s argument that “these are real people we’re talking about, they shouldn’t be subdivided into control groups and experiment zones” is wrong-headed. It’s precisely because there are real people involved that we should study what social programs work scientifically, and experiments are by far the best way to do so. For instance, voucher experiments have shown that the gains made by participants are marginal at best. Presumably Klein is glad that we’ve conducted such tests rather than simply enacting vouchers wholesale.

To be sure, we don’t want to conduct experiments that seem likely to make people in the Gulf worse off. But there’s no reason we can’t carefully test various anti-poverty interventions that seem likely to help, and assess which ones give us the biggest bang for our buck. That’s how we can help “real people” most effectively.